Meiotic DSB repair by homologous recombination occurs via multiple processes defined by distinct decisions points. One important decision involves partner choice, between recombining with the sister chromatid (the dominant repair partner during mitosis) or with the homolog (the homologous chromosome of different parental origin, the preferred partner during meiosis). Another important decision involves recombination pathway choice, between producing crossovers, where flanking chromosome sequences are exchanged, or noncrossovers. A signature contribution of our group was the demonstration that crossover and noncrossover recombination proceed via different mechanisms that diverge after initial stages of strand invasion, and that feature different biochemical activities and genetic requirements. Work during previous review periods had shown that the conserved Sgs1-Top3-Rmi1 helicase-topoisomerase complex (STR) is responsible for partitioning early recombination events between noncrossover and crossover pathways. Sgs1-Top3-Rmi1 is the yeast homolog of the mammalian BLM helicase-Top3alpha-BLAP75 complex, implicated in cancer avoidance and recombination control in humans. We showed that all three members of the yeast complex are essential for normal recombination partner choice and for population of regulated meiotic crossover and noncrossover recombination pathways. Based on these findings, we hypothesized that STR, by promoting frequent disassembly of early strand invasion intermediates, acts as a chaperone for early recombination intermediates. We hypothesized that these repeated cycles of strand invasion and disassembly would result in template switching, which in turn would lead to recombinants with mosaic parental strand contributions. This hypothesis has now been confirmed by high-throughput sequencing of recombinants that occur in a highly polymorphic test interval; more than 2/3 of recombinants display clear evidence for template switching, multiple strand invasions, or both. In addition, we uncovered evidence for other activities, including branch migration and exonucleolytic gap-formation, that appear to be specific to the crossover recombination pathway. Current work is aimed at determining the proteins responsible for these activities. In addition, prompted by observations that mutants in the conserved Smc5-Smc6 DNA repair complex, specifically in its SUMO-ligase activity, show meiotic recombination defects similar to STR complex mutants, we are testing the possibility that Smc5-Smc6-mediated SUMOylation of STR proteins is important for their meiotic activity. Studies of meiotic recombination in mutants lacking STR components also identified an Sgs1-independent role for Top3-Rmi1, in resolving strand entanglements that accumulate in recombination intermediates. We have now shown, using a procedure call return-to-growth, where recombination intermediates are accumulated during wild-type meiosis and then resolved in a subsequent mitotic cell cycle, that Sgs1 activity is not required to fully resolve recombination intermediates, but Top3-Rmi1 is. In the absence of Top3-Rmi1, the unresolvable recombination intermediates that remain induce a DNA damage checkpoint which arrests cell cycle progression; in checkpoint-defective mutants, cells progress through division, but are unable to segregate chromosomes properly. These findings confirm conclusions based on meiotic studies, and also provide a way to understand how DNA damage checkpoint defects can promote chromosome mis-segregation during oncogenesis. They also raise the intriguing novel possibility that unresolved recombination intermediates may be recognized by DNA damage checkpoint proteins through previously unknown mechanisms. Impact of chromosome structure on recombination biochemistry. Finally, we are studying how chromosome structure, specifically the meiotic chromosome axis, contributes to the regulation of recombination. A subset of axis components, important for meiotic DSB formation and partner choice, are enriched in some regions of the genome relative to others. Using a combination of artificially-induced site-specific double-strand breaks and strategies to alter meiotic axis protein enrichment as specific loci, we are examining the contributions of these proteins to all steps in meiotic recombination, from initiation to recombination intermediate resolution.